Systemic lupus erythematosus (SLE) is a common multisystem autoimmune disease that is estimated to affect more than 500,000 Americans. Our goal is to gain a better understanding of the mechanisms that normally prevent expression of the self-reactive antibodies that cause the disease as well as to better understand where and how these mechanisms fail. Research into fundamental causes of disease in humans is often impossible due to ethical considerations. Decades of medical research attest to the usefulness of mouse models as a means to identify root causes as well as new avenues for treatment and prevention. The presence of the sle1, sle2, and/or sle3/5 NZM2410 lupus loci in C57BL/6 promotes development of an autoantibody driven disorder that shares many of the features of human SLE. It is our hypothesis that failure to properly regulate one specific part of the antibody, CDR-H3, in susceptible individuals facilitates production of anti-DNA antibodies and triggers disease. CDR-H3 is important because it lies at the very center of the antigen binding site. CDR-H3 is created de novo by VDJ joining. We propose that the sequence of the DH has a dominant effect on CDR-H3 content and that a primary mechanism normally constraining CDR-H3 is conservation of DH sequence by reading frame. We propose that autoimmunity results from failure to regulate use of the amino acids that can be found in disfavored reading frames. These amino acids can also be introduced by N addition or somatic mutation. We hypothesize that sle1, sle2, and sle3/5 genes adversely affect regulation of B cell and repertoire development by permitting increased survival of B cells bearing disfavored CDR-H3. To test these hypotheses we will breed three different DH alleles, each limited to a single DH gene segment, into the C57BL/6 genome. The first DH allele will substitute arginine and other positively charged amino acids from inverted reading frame 1 for the tyrosine and glycine that are normally encoded by the preferred reading frame 1 by deletion. The second will contain a single, normal DH as a control for the loss of the rest of the DH locus. The third will control for the loss of tyrosine and glycine content by substituting codons from hydrophobic reading frame 2. If our hypotheses are correct, enrichment for arginine in DH reading frame 1 will enhance expression of disfavored CDR-H3 sequence. This heightened expression will exceed the threshold for normal control of CDR-H3 content, accelerating expression of self-reactive antibodies, including anti-DNA. In the presence of sle1, sle2, and sle3/5, mice containing altered DH alleles will display altered patterns of B cell and repertoire development, evidence altered immune responses to antigen, demonstrate enhanced expression of IgG anti-dsDNA antibodies, and lead to accelerated expression of autoimmune disease.